| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-Id: <1455802758-13550-1-git-send-email-edumazet@google.com>
Date: Thu, 18 Feb 2016 05:39:18 -0800
From: Eric Dumazet <edumazet@...gle.com>
To: "David S . Miller" <davem@...emloft.net>
Cc: netdev <netdev@...r.kernel.org>, Ilya Dryomov <idryomov@...il.com>,
Eric Dumazet <edumazet@...gle.com>,
Eric Dumazet <eric.dumazet@...il.com>
Subject: [PATCH net] tcp/dccp: fix another race at listener dismantle
Ilya reported following lockdep splat:
kernel: =========================
kernel: [ BUG: held lock freed! ]
kernel: 4.5.0-rc1-ceph-00026-g5e0a311 #1 Not tainted
kernel: -------------------------
kernel: swapper/5/0 is freeing memory
ffff880035c9d200-ffff880035c9dbff, with a lock still held there!
kernel: (&(&queue->rskq_lock)->rlock){+.-...}, at:
[<ffffffff816f6a88>] inet_csk_reqsk_queue_add+0x28/0xa0
kernel: 4 locks held by swapper/5/0:
kernel: #0: (rcu_read_lock){......}, at: [<ffffffff8169ef6b>]
netif_receive_skb_internal+0x4b/0x1f0
kernel: #1: (rcu_read_lock){......}, at: [<ffffffff816e977f>]
ip_local_deliver_finish+0x3f/0x380
kernel: #2: (slock-AF_INET){+.-...}, at: [<ffffffff81685ffb>]
sk_clone_lock+0x19b/0x440
kernel: #3: (&(&queue->rskq_lock)->rlock){+.-...}, at:
[<ffffffff816f6a88>] inet_csk_reqsk_queue_add+0x28/0xa0
To properly fix this issue, inet_csk_reqsk_queue_add() needs
to return to its callers if the child as been queued
into accept queue.
We also need to make sure listener is still there before
calling sk->sk_data_ready(), by holding a reference on it,
since the reference carried by the child can disappear as
soon as the child is put on accept queue.
Reported-by: Ilya Dryomov <idryomov@...il.com>
Fixes: ebb516af60e1 ("tcp/dccp: fix race at listener dismantle phase")
Signed-off-by: Eric Dumazet <edumazet@...gle.com>
---
include/net/inet_connection_sock.h | 5 +++--
net/dccp/ipv4.c | 14 +++++++-------
net/dccp/ipv6.c | 14 +++++++-------
net/ipv4/inet_connection_sock.c | 14 +++++++-------
net/ipv4/tcp_ipv4.c | 14 +++++++-------
net/ipv6/tcp_ipv6.c | 14 +++++++-------
6 files changed, 38 insertions(+), 37 deletions(-)
diff --git a/include/net/inet_connection_sock.h b/include/net/inet_connection_sock.h
index 481fe1c9044c..49dcad4fe99e 100644
--- a/include/net/inet_connection_sock.h
+++ b/include/net/inet_connection_sock.h
@@ -270,8 +270,9 @@ struct dst_entry *inet_csk_route_child_sock(const struct sock *sk,
struct sock *newsk,
const struct request_sock *req);
-void inet_csk_reqsk_queue_add(struct sock *sk, struct request_sock *req,
- struct sock *child);
+struct sock *inet_csk_reqsk_queue_add(struct sock *sk,
+ struct request_sock *req,
+ struct sock *child);
void inet_csk_reqsk_queue_hash_add(struct sock *sk, struct request_sock *req,
unsigned long timeout);
struct sock *inet_csk_complete_hashdance(struct sock *sk, struct sock *child,
diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c
index 5684e14932bd..902d606324a0 100644
--- a/net/dccp/ipv4.c
+++ b/net/dccp/ipv4.c
@@ -824,26 +824,26 @@ lookup:
if (sk->sk_state == DCCP_NEW_SYN_RECV) {
struct request_sock *req = inet_reqsk(sk);
- struct sock *nsk = NULL;
+ struct sock *nsk;
sk = req->rsk_listener;
- if (likely(sk->sk_state == DCCP_LISTEN)) {
- nsk = dccp_check_req(sk, skb, req);
- } else {
+ if (unlikely(sk->sk_state != DCCP_LISTEN)) {
inet_csk_reqsk_queue_drop_and_put(sk, req);
goto lookup;
}
+ sock_hold(sk);
+ nsk = dccp_check_req(sk, skb, req);
if (!nsk) {
reqsk_put(req);
- goto discard_it;
+ goto discard_and_relse;
}
if (nsk == sk) {
- sock_hold(sk);
reqsk_put(req);
} else if (dccp_child_process(sk, nsk, skb)) {
dccp_v4_ctl_send_reset(sk, skb);
- goto discard_it;
+ goto discard_and_relse;
} else {
+ sock_put(sk);
return 0;
}
}
diff --git a/net/dccp/ipv6.c b/net/dccp/ipv6.c
index 9c6d0508e63a..b8608b71a66d 100644
--- a/net/dccp/ipv6.c
+++ b/net/dccp/ipv6.c
@@ -691,26 +691,26 @@ lookup:
if (sk->sk_state == DCCP_NEW_SYN_RECV) {
struct request_sock *req = inet_reqsk(sk);
- struct sock *nsk = NULL;
+ struct sock *nsk;
sk = req->rsk_listener;
- if (likely(sk->sk_state == DCCP_LISTEN)) {
- nsk = dccp_check_req(sk, skb, req);
- } else {
+ if (unlikely(sk->sk_state != DCCP_LISTEN)) {
inet_csk_reqsk_queue_drop_and_put(sk, req);
goto lookup;
}
+ sock_hold(sk);
+ nsk = dccp_check_req(sk, skb, req);
if (!nsk) {
reqsk_put(req);
- goto discard_it;
+ goto discard_and_relse;
}
if (nsk == sk) {
- sock_hold(sk);
reqsk_put(req);
} else if (dccp_child_process(sk, nsk, skb)) {
dccp_v6_ctl_send_reset(sk, skb);
- goto discard_it;
+ goto discard_and_relse;
} else {
+ sock_put(sk);
return 0;
}
}
diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c
index 46b9c887bede..64148914803a 100644
--- a/net/ipv4/inet_connection_sock.c
+++ b/net/ipv4/inet_connection_sock.c
@@ -789,14 +789,16 @@ static void inet_child_forget(struct sock *sk, struct request_sock *req,
reqsk_put(req);
}
-void inet_csk_reqsk_queue_add(struct sock *sk, struct request_sock *req,
- struct sock *child)
+struct sock *inet_csk_reqsk_queue_add(struct sock *sk,
+ struct request_sock *req,
+ struct sock *child)
{
struct request_sock_queue *queue = &inet_csk(sk)->icsk_accept_queue;
spin_lock(&queue->rskq_lock);
if (unlikely(sk->sk_state != TCP_LISTEN)) {
inet_child_forget(sk, req, child);
+ child = NULL;
} else {
req->sk = child;
req->dl_next = NULL;
@@ -808,6 +810,7 @@ void inet_csk_reqsk_queue_add(struct sock *sk, struct request_sock *req,
sk_acceptq_added(sk);
}
spin_unlock(&queue->rskq_lock);
+ return child;
}
EXPORT_SYMBOL(inet_csk_reqsk_queue_add);
@@ -817,11 +820,8 @@ struct sock *inet_csk_complete_hashdance(struct sock *sk, struct sock *child,
if (own_req) {
inet_csk_reqsk_queue_drop(sk, req);
reqsk_queue_removed(&inet_csk(sk)->icsk_accept_queue, req);
- inet_csk_reqsk_queue_add(sk, req, child);
- /* Warning: caller must not call reqsk_put(req);
- * child stole last reference on it.
- */
- return child;
+ if (inet_csk_reqsk_queue_add(sk, req, child))
+ return child;
}
/* Too bad, another child took ownership of the request, undo. */
bh_unlock_sock(child);
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index c84477949d3a..487ac67059e2 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -1597,30 +1597,30 @@ process:
if (sk->sk_state == TCP_NEW_SYN_RECV) {
struct request_sock *req = inet_reqsk(sk);
- struct sock *nsk = NULL;
+ struct sock *nsk;
sk = req->rsk_listener;
if (unlikely(tcp_v4_inbound_md5_hash(sk, skb))) {
reqsk_put(req);
goto discard_it;
}
- if (likely(sk->sk_state == TCP_LISTEN)) {
- nsk = tcp_check_req(sk, skb, req, false);
- } else {
+ if (unlikely(sk->sk_state != TCP_LISTEN)) {
inet_csk_reqsk_queue_drop_and_put(sk, req);
goto lookup;
}
+ sock_hold(sk);
+ nsk = tcp_check_req(sk, skb, req, false);
if (!nsk) {
reqsk_put(req);
- goto discard_it;
+ goto discard_and_relse;
}
if (nsk == sk) {
- sock_hold(sk);
reqsk_put(req);
} else if (tcp_child_process(sk, nsk, skb)) {
tcp_v4_send_reset(nsk, skb);
- goto discard_it;
+ goto discard_and_relse;
} else {
+ sock_put(sk);
return 0;
}
}
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index 1a5a70fb8551..5c8c84273028 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -1387,7 +1387,7 @@ process:
if (sk->sk_state == TCP_NEW_SYN_RECV) {
struct request_sock *req = inet_reqsk(sk);
- struct sock *nsk = NULL;
+ struct sock *nsk;
sk = req->rsk_listener;
tcp_v6_fill_cb(skb, hdr, th);
@@ -1395,24 +1395,24 @@ process:
reqsk_put(req);
goto discard_it;
}
- if (likely(sk->sk_state == TCP_LISTEN)) {
- nsk = tcp_check_req(sk, skb, req, false);
- } else {
+ if (unlikely(sk->sk_state != TCP_LISTEN)) {
inet_csk_reqsk_queue_drop_and_put(sk, req);
goto lookup;
}
+ sock_hold(sk);
+ nsk = tcp_check_req(sk, skb, req, false);
if (!nsk) {
reqsk_put(req);
- goto discard_it;
+ goto discard_and_relse;
}
if (nsk == sk) {
- sock_hold(sk);
reqsk_put(req);
tcp_v6_restore_cb(skb);
} else if (tcp_child_process(sk, nsk, skb)) {
tcp_v6_send_reset(nsk, skb);
- goto discard_it;
+ goto discard_and_relse;
} else {
+ sock_put(sk);
return 0;
}
}
--
2.7.0.rc3.207.g0ac5344
Powered by blists - more mailing lists